torch_em.data.datasets.medical.piccolo

  1"""The PICCOLO dataset contains annotations for polyp segmentation
  2in narrow band imaging colonoscopy.
  3
  4NOTE: Automatic download is not supported with this dataset. See 'get_piccolo_data' for details.
  5
  6The dataset is from the publication https://doi.org/10.3390/app10238501.
  7Please cite it if you use this dataset for your research.
  8"""
  9
 10import os
 11from glob import glob
 12from natsort import natsorted
 13from typing import Union, Tuple, Literal, List
 14
 15from torch.utils.data import Dataset, DataLoader
 16
 17import torch_em
 18
 19from .. import util
 20
 21
 22def get_piccolo_data(path: Union[os.PathLike, str], download: bool = False) -> str:
 23    """Get the PICCOLO dataset.
 24
 25    The database is located at:
 26    - https://www.biobancovasco.bioef.eus/en/Sample-and-data-e-catalog/Databases/PD178-PICCOLO-EN1.html
 27
 28    Follow the instructions below to get access to the dataset.
 29    - Visit the attached website above
 30    - Fill up the access request form: https://labur.eus/EzJUN
 31    - Send an email to Basque Biobank at solicitudes.biobancovasco@bioef.eus, requesting access to the dataset.
 32    - The team will request you to follow-up with some formalities.
 33    - Then, you will gain access to the ".rar" file.
 34    - Finally, provide the path where the rar file is stored, and you should be good to go.
 35
 36    Args:
 37        path: Filepath to a folder where the data is downloaded for further processing.
 38        download: Whether to download the data if it is not present.
 39
 40    Returns:
 41        Filepath where the data is downloaded.
 42    """
 43    data_dir = os.path.join(path, r"piccolo dataset-release0.1")
 44    if os.path.exists(data_dir):
 45        return data_dir
 46
 47    if download:
 48        raise NotImplementedError(
 49            "Automatic download is not possible for this dataset. See 'get_piccolo_data' for details."
 50        )
 51
 52    rar_file = os.path.join(path, r"piccolo dataset_widefield-release0.1.rar")
 53    if not os.path.exists(rar_file):
 54        raise FileNotFoundError(
 55            "You must download the PICCOLO dataset from the Basque Biobank, see 'get_piccolo_data' for details."
 56        )
 57
 58    util.unzip_rarfile(rar_path=rar_file, dst=path, remove=False)
 59    return data_dir
 60
 61
 62def get_piccolo_paths(
 63    path: Union[os.PathLike, str], split: Literal['train', 'validation', 'test'], download: bool = False
 64) -> Tuple[List[str], List[str]]:
 65    """Get paths to the PICCOLO data.
 66
 67    Args:
 68        path: Filepath to a folder where the data is downloaded for further processing.
 69        split: The choice of data split.
 70        download: Whether to download the data if it is not present.
 71
 72    Returns:
 73        List of filepaths for the image data.
 74        List of filepaths for the label data.
 75    """
 76    data_dir = get_piccolo_data(path, download)
 77
 78    image_paths = natsorted(glob(os.path.join(data_dir, split, "polyps", "*")))
 79    gt_paths = natsorted(glob(os.path.join(data_dir, split, "masks", "*")))
 80
 81    return image_paths, gt_paths
 82
 83
 84def get_piccolo_dataset(
 85    path: Union[os.PathLike, str],
 86    patch_shape: Tuple[int, int],
 87    split: Literal["train", "validation", "test"],
 88    resize_inputs: bool = False,
 89    download: bool = False,
 90    **kwargs
 91) -> Dataset:
 92    """Get the PICCOLO dataset for polyp segmentation in narrow band imaging colonoscopy images.
 93
 94    Args:
 95        path: Filepath to a folder where the data is downloaded for further processing.
 96        patch_shape: The patch shape to use for training.
 97        split: The choice of data split.
 98        resize_inputs: Whether to resize inputs to the desired patch shape.
 99        download: Whether to download the data if it is not present.
100        kwargs: Additional keyword arguments for `torch_em.default_segmentation_dataset`.
101
102    Returns:
103        The segmentation dataset.
104    """
105    image_paths, gt_paths = get_piccolo_paths(path, split, download)
106
107    if resize_inputs:
108        resize_kwargs = {"patch_shape": patch_shape, "is_rgb": True}
109        kwargs, patch_shape = util.update_kwargs_for_resize_trafo(
110            kwargs=kwargs, patch_shape=patch_shape, resize_inputs=resize_inputs, resize_kwargs=resize_kwargs
111        )
112
113    return torch_em.default_segmentation_dataset(
114        raw_paths=image_paths,
115        raw_key=None,
116        label_paths=gt_paths,
117        label_key=None,
118        patch_shape=patch_shape,
119        is_seg_dataset=False,
120        **kwargs
121    )
122
123
124def get_piccolo_loader(
125    path: Union[os.PathLike, str],
126    batch_size: int,
127    patch_shape: Tuple[int, int],
128    split: Literal["train", "validation", "test"],
129    resize_inputs: bool = False,
130    download: bool = False,
131    **kwargs
132) -> DataLoader:
133    """Get the PICCOLO dataloader for polyp segmentation in narrow band imaging colonoscopy images.
134
135    Args:
136        path: Filepath to a folder where the data is downloaded for further processing.
137        batch_size: The batch size for training.
138        patch_shape: The patch shape to use for training.
139        split: The choice of data split.
140        resize_inputs: Whether to resize inputs to the desired patch shape.
141        download: Whether to download the data if it is not present.
142        kwargs: Additional keyword arguments for `torch_em.default_segmentation_dataset` or for the PyTorch DataLoader.
143
144    Returns:
145        The DataLoader.
146    """
147    ds_kwargs, loader_kwargs = util.split_kwargs(torch_em.default_segmentation_dataset, **kwargs)
148    dataset = get_piccolo_dataset(path, patch_shape, split, resize_inputs, download, **ds_kwargs)
149    return torch_em.get_data_loader(dataset, batch_size, **loader_kwargs)